A camera detects a face portion of a subject from a shot image to determine whether a shot image is a good shot candidate image from the features of the face portion and a face portion-related image portion. As examples, the features include, but not be limited to, (1) the shape and variation in contour ranging from the vicinity of the face portion, (2) the position of a hand or leg to be determined by similarity to the skin tone of the face portion, and (3) determination made from the detection result of the face portion as to whether the subject is a half-length or full-length figure.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A camera comprising: a face portion detection part configured to detect, in an image signal from an image pickup device, a face portion included in the image signal; a face position detection part configured to detect a position and a size of the detected face portion; a whole subject image determining part configured to determine whether the image is a whole subject image or a half subject image based on the detected position and largeness of the face portion; a pose determining part configured to, (1) if the image is determined as a half subject image, determine a pose of the subject based on whether a positional relationship between the face position and a position corresponding to a hand is a predetermined positional relationship, and (2) if the image is determined as a whole subject image, determine a pose of the subject based on whether a positional relationship between the face position and a position corresponding to a leg is a predetermined positional relationship.
2. The camera according to claim 1 further comprising: a skin color detection part configured to detect a color of the face portion as a skin color; and a hand portion determining part configured to, if the whole subject image determining part determines that the image is a half image of the subject, determine a part corresponding to a hand based on a color detected by the skin color detection part.
3. The camera according to claim 1 further comprising: a skin color detection part configured to detect a color of the face portion as a skin color; and a leg portion determining part configured to, if the whole subject image determining part determines that the image is a whole image of the subject, determine a part corresponding to a leg based on the color detected by the skin color detection part.
4. The camera according to claim 1 wherein the pose detection part determines a pose of the subject based on a recorded coordinates of a face and at least one of a hand and a leg, and the detected face position and at least one of the detected hand position and the detected leg position.
5. The camera according to claim 1 wherein the whole subject image determining part determines the image as a whole subject image when yA is more than or equal to zero and as a half subject image when yA is less than zero, where yA is defined in a formula: yA=yk+(Δy/2)−6·Δy where, by setting coordinates as such a Y coordinate is set on a direction of a leg, yk is a position of a face, Δy is a largeness of the face, and yA is a position of a leg.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
January 31, 2012
July 22, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.